We are currently investigating, why our cluster isnt load balancing to nodes with lower cpu/ram consumption.
After fixing, that the 2 Metrics "servicefabric:/_MemoryInMB” and “servicefabric:/_CpuCores" were not showing up, (ServiceFabric: Missing default metrics "servicefabric:/_MemoryInMB" and "servicefabric:/_CpuCores") now the 2 metrics always show 0% usage on all nodes. It feels like the cluster isnt reporting the metric at all. But from what i understand, developer aren't supposed to manually report those metrics. (Especially because its hard to measure them)
So... Im kind of clueless here. As far as i know the metrics used so be capital letters ("Servicefabric:/ ...." instead of "servicefabric:/") and they just recently fixed that service fabric isnt recognizing the metric because of the upper/lowercase difference. (look at the link for more info)
Could it be the case that they dont automatically report to the metrics that are lower cased?
Thanks in advance!
Resource governance limits such as CpuCores or MemoryInMB are specified in the application manifest. Please see https://learn.microsoft.com/en-us/azure/service-fabric/service-fabric-resource-governance